Ahmadreza Samadi / HomeEssay · Start with the decision
Start with the decision, then build the AI
An AI product earns its place when it helps someone make a useful decision and act on it. Define the evidence, responsible person and acceptance rule before choosing a model, then test the complete workflow.
| Step | Value or evidence | Interpretation |
|---|---|---|
| Decision | Which fictional records need investigation? | A named researcher chooses the shortlist |
| Evidence | Dated sources with missing fields visible | Unresolved claims remain unresolved |
| Acceptance | Reviewer checks the proposed research note | Generated text alone is not completion |
A decision brief before a model brief
Define the decision, evidence, responsible person and acceptance rule before choosing the AI system. A useful experiment tests the whole path from evidence to action.
For the hypothetical property shortlist in this essay, write the brief as: “A researcher chooses which records deserve further investigation, using dated evidence with unresolved fields visible.” Compare that process with the current manual approach. A shortlist is a research aid; it is not a purchase recommendation or a verified finding about a property.
A checklist for the next experiment
- Name one decision and the person responsible for it.
- Record the current process, including its cost and failure cases.
- Define acceptable evidence and what happens when it is missing.
- Set the acceptance rule and the human review point before testing.
- Compare the same task under both workflows; retain rejected cases and review effort.
Use the decision brief or tool · Explore the three signature ideas
Primary references
- Google: Rules of Machine Learning
- NIST AI RMF 1.0
- Microsoft Research: Guidelines for Human-AI Interaction
These references inform the evaluation and interaction principles. The workflow above is an editorial synthesis, not an endorsed standard or a reported company outcome.
When I assess an AI product idea, I start with a business question: what decision is difficult today, and what would make it easier?
That question provides a practical starting point for both product strategy and engineering. It identifies who needs the result, what information matters, and what the system must do beyond producing an answer.
Which decision should this AI product help someone make?
Consider a hypothetical property research tool. “Analyze properties with AI” leaves many product choices unresolved. “Help an investor decide which properties deserve further investigation” is more specific. It identifies an action and a point in a workflow where better information could help.
The distinction matters because a useful research tool may need to show supporting evidence, make comparisons easy, and identify missing information. A single score would only cover part of that job.
What evidence would make the decision responsible?
Once the decision is clear, ask what evidence a person would need to make it responsibly. In this example, that might include imagery, its capture date, relevant property attributes, and the limits of what can be inferred.
Then ask whether the necessary data is available, sufficiently current, and appropriate for the use case. These questions belong early in product development because the answers can change what the product should promise.
What action follows the answer?
An output becomes useful when it fits the next step. Does the customer shortlist an opportunity, request additional information, or send a case for human review?
That action shapes the interface and the architecture. It determines which information must be retained, which integrations matter, and where a person needs to be able to question or correct the system.
How would I evaluate the whole workflow?
For this hypothetical tool, I would evaluate several questions together:
- Does it surface information that is relevant to the research decision?
- Can a user understand the evidence and its limitations?
- How much review is needed before taking the next step?
- What happens when information is incomplete or the system is wrong?
- Is the value of that assistance worth the cost of delivering it?
A model metric can inform those questions, but it cannot answer all of them. Product evaluation also needs to examine the experience of the person using the result.
What is the smallest useful commitment?
I would rather define an initial product around one decision we can support well than make a broad promise that is difficult to evaluate. A focused scope gives the team a clearer way to learn from customers and decide what to build next.
My approach is to connect the customer’s decision, the supporting evidence, the product workflow, and the engineering effort. That connection is where technical capability becomes a business proposition.
Put the method to work
Before You Build the AI turns these questions into an interactive decision brief. Explore a worked example, adapt it to a product idea, and export a first experiment.
For the professional context behind my interest in property research workflows, see my work on AI real estate intelligence at Appraiva. The example in this essay remains hypothetical.
Primary sources and scope
These three primary sources were opened on 20 October 2026. Access dates do not establish a new source publication date.
- Google: Rules of Machine Learning. Living engineering guide; no new publication date asserted.
- NIST: AI Risk Management Framework. AI RMF 1.0 released January 2023.
- Microsoft Research: Guidelines for Human-AI Interaction. Research guidelines introduced in 2019.
The examples and proposed methods are editorial teaching work. They are not company results or independently certified findings. Educational, not financial advice.
Prefer a copy you can edit in a text editor? Download the decision-first text kit (in English) for a blank template, a separate fictional example and reading notes. Extract the ZIP and save a private working copy. This is another format for the existing planning aid; no experiment was run.